Table 2.

Results from unadjusted models with individual sleep health dimensions and the risk of heart disease.

Variable	B	SE	p	95% CI
Model 1. self-report model
Regularity1	− 0.08	0.03	.004	− 0.14, − 0.03
Satisfaction	0.14	0.03	<.001	0.08, 0.20
Alertness	0.19	0.02	<.001	0.14, 0.23
Efficiency	0.10	0.02	<.001	0.06, 0.15
Duration	− 0.06	0.03	.058	− 0.12, 0.002
Model 2. actigraphy/self-report model
Regularity1	0.03	0.09	.779	− 0.15, 0.21
Satisfaction	0.17	0.09	.045	0.004, 0.34
Alertness	0.04	0.10	.693	− 0.16, 0.23
Timing	0.22	0.06	<.001	0.11, 0.33
Efficiency	0.10	0.11	.360	− 0.11, 0.31
Duration	− 0.02	0.13	.862	− 0.28, 0.23

¹For each sleep health dimension, higher scores indicated poorer sleep (i.e., irregularity, poorer satisfaction, lack of alertness, inefficiency, later midpoint timing, and shorter duration). All sleep dimensions were treated continuously and z-scored, then entered simultaneously in the same model. The beta coefficients from this table were used to create the weighted-regression sleep health composites:

Self-Report Weighted Sleep Health Composite $=(-0.08\ast \mathrm{Regularity})+(0.14\ast \mathrm{Satisfaction})+(0.19\ast \mathrm{Alertness})+(0.10\ast \mathrm{Efficiency})+(-0.06\ast \mathrm{Duration})$.

Actigraphy/self-report Weighted Sleep Heath Composite $=(0.03\ast \mathrm{Regularity})+(0.17\ast \mathrm{Satisfaction})+(0.04\ast \mathrm{Alertness})+\left(0.22,,\ast ,\mathrm{Timing}\right)+(0.10\ast \mathrm{Efficiency})+(-0.02\ast \mathrm{Duration})$.